WO2022218706A1

WO2022218706A1 - Method for analysing the manufacturability of products

Info

Publication number: WO2022218706A1
Application number: PCT/EP2022/058412
Authority: WO
Inventors: Irlan Grangel Gonzalez; Felix LOESCH
Original assignee: Robert Bosch Gmbh
Priority date: 2021-04-13
Filing date: 2022-03-30
Publication date: 2022-10-20
Also published as: DE102021109170A1

Abstract

One aspect of the present disclosure relates to a computer-implemented method for analysing the manufacturability of products. The method comprises providing a computer-implemented machine knowledge graph containing information about a multiplicity of machines. The machine knowledge graph comprises a conceptual model that comprises a multiplicity of concepts, wherein each concept maps a corresponding machine capability from a multiplicity of machine capabilities, wherein the concepts are provided with attributes and their respective relationships among one another are defined. The machine knowledge graph of the first aspect furthermore comprises a multiplicity of machine entities that are assigned to the respective concepts in the conceptual model. Each machine entity contains data points of machine capabilities of the corresponding machine from the multiplicity of machines. The method furthermore comprises providing one or more features of a product to be manufactured in the machine knowledge graph. Finally, the method comprises ascertaining the manufacturability of a product based on the information contained in the machine knowledge graph.

Description

METHOD OF ANALYZING A MANUFACTURER OF

PRODUCTS

technical field

The present invention relates to techniques for analyzing manufacturability of products. Related aspects relate to a computer program and a computer-implemented system.

background

A typical problem when developing new products or product variants is to carry out a manufacturability analysis of products or product variants on one or more machines, e.g. B. on one or more industrial machines in a specific industrial plant. This task can be understood as a simulation process that takes into account the technical characteristics of a new product and the technical capabilities of one or more available machines to determine whether the one or more machines can produce the new product or product variant. In some prior art methods, this task is addressed by a time-consuming and error-prone analysis that is performed partially manually. For example, production engineers perform the analysis by consulting a large number of different data sources and especially expert knowledge to determine whether the new product or product variant can be manufactured on a specific machine. In practice, this can lead to a great deal of effort in answering manufacturability questions. The main reason for the high effort is that the data required for the decision on the manufacturability is usually missing in some prior art methods or is not available in a form that allows automated processing using computer-implemented systems. Furthermore, the data can be distributed across different computer-implemented systems that communicate little or not at all, which can lead to the introduction of semantic heterogeneity conflicts (e.g. naming the same entities in different ways). Some prior art techniques are due In view of the increasing variety of products and the complexity of machines, it is not efficient in deciding on the manufacturability of products or product variants, because complex calculations may be required to determine whether a specific machine or a specific group of machines is suitable for the manufacture of the product under consideration is.

Therefore, there is a need to develop new efficient and especially automated techniques that can solve some or all of the above problems.

Summary of the Invention

A first general aspect of the present disclosure relates to a computer-implemented method for analyzing a manufacturability of products. The method includes providing a computer implemented machine knowledge graph containing information about a plurality of machines. The machine knowledge graph comprises a conceptual model comprising a plurality of concepts, each concept mapping a corresponding one of a plurality of machine capabilities, the concepts being attributed and their respective interrelationships being defined. The machine knowledge graph of the first aspect further includes a plurality of machine instances associated with the respective concepts in the conceptual model. Each machine instance comprises machine capability data points of the corresponding machine in the plurality of machines. The method also includes providing one or more features of a product to be manufactured in the machine knowledge graphs. Finally, the method includes determining the manufacturability of a product based on the information contained in the machine knowledge graph.

A second general aspect of the present disclosure relates to a computer program that is designed to carry out the computer-implemented method according to the first general aspect of the present disclosure. A third general aspect of the present disclosure relates to a computer-implemented system for analyzing a manufacturability of products, configured to execute the computer program according to the second general aspect of the present disclosure.

The techniques of the first through third general aspects may have one or more of the following advantages.

First, the techniques of the present disclosure provide the ability to provide a method for analyzing a manufacturability of products using expert knowledge about product features and machine capabilities contained in a knowledge graph such that analyzing manufacturability is performed automatically on a computer-implemented system instead of using a lengthy process that requires analyzing and communicating between many different systems. Thus, with the present techniques, the outlay for analyzing manufacturability can be significantly reduced. In other words, a simulation can be run to determine if the technical capabilities of a particular machine are sufficient to produce a particular product or product variant.

Secondly, the present techniques can make it possible to efficiently find out on which machine or group of machines a certain product with certain characteristics can be manufactured. In addition, the present techniques can find out the reasons why it is not possible to manufacture a certain product on a given set of machines by performing the appropriate calculations. Based on these calculations, the techniques of the present disclosure may suggest how certain machines may need to be reconfigured to produce additional product variants, resulting in higher utilization of existing manufacturing assets (eg, machinery present in a particular industrial facility). In addition, the present techniques can provide valuable information to engineers on how to redesign product variants at an early stage so that they can be manufactured on existing manufacturing facilities. Some terms are used in the present disclosure in the following manner:

The term "product" includes any product or good that is manufactured in an industrial way (e.g. in an automated or partially automated manufacturing process) on one or more (industrial) machines. The product can be an electrical or electro-mechanical component, for example. The product may be a finished device (designed to perform a specific function). In other examples, the product may be an intermediate stage of a manufacturing process designed for further processing. In some cases, the product can be a computer, a tablet computer, a smartphone or an electronic device (e.g. a tool, a household appliance or a gardening device). In other examples, the product may be a vehicle component (the term "vehicle" is understood to mean any device designed to transport passengers and/or cargo, for example the vehicle may be a motor vehicle, in particular an at least partially autonomously operating / assisted motor vehicle) . In still other examples, a product may be an apparatus for use in an industrial environment (e.g., an assembly line). The term "product" also includes all internal (e.g. mechanical or electrical) components or parts of the above devices or devices (e.g. a motor, a control device, a gearbox, a camera-based system, a motor control, an assistance system, a sensor , a lamp, a battery, a circuit board, a housing). An internal component (or part) can also be part of the systems described above or a combination of several of the systems described above (or parts of them).

The term "product variant" includes a modification of a product with regard to one or more technical or other characteristics (e.g. dimensions, materials used, etc.).

Accordingly, the term "product features" includes all features that define a product, its components or parts (e.g., their dimensions, shape, weight, or material, etc.) and which should be taken into account when manufacturing the product.

The term "machine" means any industrial machine that is suitable for manufacturing a product. The machine of the present disclosure may be a fully automated or semi-automated industrial machine controlled, for example, by a system such as a computer. The term "machine" also includes a group of industrial machines (e.g. an industrial plant or production line with linked industrial machines), which may be connected in a single manufacturing process. For example, the industrial machine or group of industrial machines can be designed to manufacture different components of a device.

Accordingly, the term "machine capability" includes any property of a machine, for example an industrial machine as explained above, which plays a role in the manufacturability of a product (e.g. workspace, existing tools and/or capabilities of the existing tools or more complex information that is necessary for the manufacturability of a product or a product component play a role).

“Manufacture” in the present disclosure includes both processes in which material is formed or otherwise modified (e.g., an injection molding process) and processes in which components are connected or assembled.

In the following, the term "knowledge graph" refers to all approaches that describe data points (e.g. data that depict the respective product features of a specific product or machine capabilities of a specific machine or machine type) on a semantic level, i.e. approaches in which the data points even be transferred to conceptual models in order to make them usable. The data points can be interpreted as data instances of concepts, which are defined in a conceptual model with appropriate relationships to each other. In addition, attributes can be assigned to the concepts. A knowledge graph consists of its conceptual model (or ontology in others terminology) and the data instances of the concepts defined in the conceptual model. Further explanations can be found below.

Brief description of the figures

1a is a flow chart illustrating an example of a computer-implemented method for analyzing a manufacturability of products according to the first aspect. 1b and 1c are flow charts showing further possible method steps according to the first aspect.

Fig. 2 shows schematically a structure of a machine knowledge graph 1 with a conceptual model 10 and a plurality of machine instances 20. A structure of a knowledge graph of product features 2 with a conceptual model of product features 30 and a plurality of product instances 40 can also be seen in this figure .

FIG. 3 schematically shows 20 possible configurations and further aspects of a computer-implemented method for analyzing the manufacturability of products.

Detailed description

First, the techniques for analyzing the manufacturability of products are described with reference to FIGS. 1a to 1c. Then, exemplary structures of the machine knowledge graph 1 and the knowledge graph of product features 2 are discussed with reference to FIG. 2 . Finally, possible configurations and further aspects of the present disclosure are presented with reference to FIG. 3 .

As outlined in FIGS. 1a to 1c, a first general aspect relates to a computer-implemented method for analyzing a manufacturability of products. Manufacturability within the meaning of the present disclosure means a possibility to manufacture a product, for example with respective product components, through a manufacturing process (eg on a fully automatic or semi-automatic industrial machine or a group of such industrial machines). The method steps of the corresponding independent claim are shown in the solid line boxes in Fig. La to lc, while the method steps of some dependent claims are shown in the boxes represented by dashed lines.

The method initially includes the provision 100 of a machine knowledge graph 1 implemented on a computer, which contains information about a plurality of machines. The term "machine knowledge graph" means that this knowledge graph refers to a machine or a type of machine, and is only used to distinguish it from other knowledge graphs in cases where multiple knowledge graphs are used. The machines can be, for example, machine tools (eg milling machines or lathes, mechanical presses and machine hammers for forging, and eroding machines) or power machines (eg electric power machines). As mentioned above, in some examples, the machine knowledge graph may include a conceptual model 10 having a plurality of concepts 11,12. In one example, each concept may map a corresponding one of a plurality of machine capabilities. In addition, the concepts can be provided with attributes 13-15 and their respective relationships with one another can be defined. A concept can be viewed as a node 11, 12 of the knowledge graph 1 and a relationship between the concepts as edges that connect the concepts (or nodes) to one another. In some examples, a machine capability may refer to a machine capability of a machine type. The machine capability can, for example, relate to a size and/or a material of the product which the machine can accommodate or process (eg in a manufacturing process). In another example, the machine capability may be a working temperature interval that the machine can provide during a manufacturing process (eg, when a product or product components need to be manufactured below a certain temperature interval, eg, refrigerated). In still other examples, machine capability may refer to functional product features that the machine can provide during the manufacturing process. For example, in some cases, the product components need to be assembled in a certain way in order to fulfill the required product functions (e.g., a case top must be tightened with the bottom with screws to prevent the two parts at high temperatures fall apart). In addition, the machine capabilities of a machine can vary from a system, such as a computer-implemented system, may be queried, and the machine may provide information regarding the machine capabilities to the system as a result of that query (eg, if the machine is operational). In another example, a plurality of machine capabilities of a machine or machine type relevant to the manufacture of a product may be defined 50 by manufacturing engineers. In an example, a machine capability of a machine may be given by a workspace for manufacture of a product (in Figure 2 the corresponding concept is denoted as "workspace" 11 ). In this context, the concept of "workspace" can be further provided with the attribute that can include a maximum workspace (for example, a minimum or maximum length, width or height of products or product components, or a combination of these characteristics) that a machine can accommodate and/ or can edit (in Fig. 2 the corresponding attribute is denoted as "value ability" 14). Furthermore, a machine capability of a machine can be a housing material that it can machine: This concept is denoted as “housing material” 12 in FIG. 2 . The concept "housing material" can in turn be provided with an attribute that, for example, provides information about a number of possible materials (e.g. metal, plastic and/or materials manufactured with a certain process, such as die casting) that a machine can process (referred to as “enumeration capability” 15 in Figure 2). The relationship between the concepts may be such that, in some examples, selected features of one concept dictate the selection of another concept (e.g., a machine may have different working areas of the "working area" concept that depend on the selected material of the "housing material" concept) . In this way, the conceptual model can include all possible machine capabilities 51 offered by a machine type or machine. In some examples, machine capabilities may be associated with machine types. Mapping machine capabilities to machine types can be used when all machines of a given machine type offer the same machine capabilities. If this is not the case, machine abilities e.g. B. can be assigned individually to specific machines. In an example, the conceptual model may contain the concepts related to the machines belonging to the same group of machines belong. In other examples, the conceptual model may include the concepts related to machines belonging to different groups of machines. For example, the groups of machines can be located in the same industrial plant or in different industrial plants.

In the present disclosure, the machine knowledge graph may include a plurality of machine instances 20 associated with the respective concepts in the conceptual model. In some examples, each machine instance 21 may include machine capability data points 22-24 of the corresponding one of the plurality of machines. In other words, a conceptual model representing general characteristics of the plurality of machine capabilities can be instantiated 52 by defining concrete machine instances of the concepts contained in the conceptual model that relate to a particular machine or machine type. In some examples, such an association between a particular machine and the corresponding concepts may be made through machine types if the machine capabilities have been mapped to the machine types. In other examples, this connection can be provided directly to a machine under consideration if the machine capabilities are directly associated with the machines. For example, as shown in Fig. 2, the machine instance of a particular machine may be given by the data points containing the machine capabilities "WorkArea1" 22, "Housing Materiall" 23 and "Housing Material2" 24, each associated with the Concepts "working area" 11 and "housing material" 12 of the conceptual model are connected. In the embodiment of FIG. 2, the data points "Working area" comprise the definition of the maximum values of the working area of a particular machine, ie length of 120 mm, width of 250 mm and height of 50 mm. In other examples, the working area can be defined by other sizes. In addition, the data points "Housing Material1" and "Housing Material2" define the housing material processed by a specific machine, in this example metal or plastic. In other examples, other materials can be used. Thereby the machine instances of concrete machines (eg some or all machines of the plurality of machines) can be stored in the knowledge graph and this can be combined in a computer-implemented system be implemented 53, 54 with the conceptual model (eg in a computer program running on the computer-implemented system).

The next step of the computer-implemented method includes providing 200 one or more features of a product to be manufactured in the machine knowledge graphs. These product features can e.g. B. the dimensions or shape of a product or product components (e.g., length, width, and height), the product material of a product or product component (e.g., metal, plastic, or die-cast), the weight of a product, or other product characteristics that describe physical (e.g. mechanical or electrical) or chemical properties of the product (or its components) that are relevant to the manufacture of the product. In some examples, a plurality of product features of a product may be defined 60 by product engineers. Finally, the manufacturability of a product can be determined 300 using the information contained in the machine knowledge graph. The machine knowledge graph 1 of the present disclosure can be configured to receive the one or more characteristics of the product. Furthermore, the machine knowledge graph 1 can be designed to transfer received features of the product into the conceptual model and/or into the data instances.

In the present technique, providing 200 the one or more features of the product to be manufactured in the machine knowledge graphs may further comprise providing 210 a knowledge graph of product features 2 . This can contain information about a number of products. Similar to that described above in connection with the machine knowledge graph 1 , in some examples the knowledge graph of product features may contain a conceptual model of product features 30 comprising a different plurality of concepts 31 , 32 . In an example, each concept from the other plurality of concepts may map a corresponding one of a plurality of product features. In addition, the concepts from the other plurality of concepts can be provided with attributes and their respective relationships with one another are defined. A concept from the other plurality of concepts can be viewed as a node 31, 32 of the knowledge graph of product features 2 and a relationship between the Concepts as edges that connect the concepts (or nodes) together. In some examples, typical product characteristics can be defined based on the requirements for the plurality of products (e.g., a product cluster). In one example, a product feature of a product may be given by a product size (in Figure 2 the corresponding concept is denoted as "product size" 31). Furthermore, a product feature of a product can include a product material (see label “product material” 32 in FIG. 2 for the corresponding concept) that is required to manufacture the product. In addition, one or more concepts of the conceptual model of product features may contain information about requirements related to product components of a product that should be assembled in a certain way to fulfill the required product functions (e.g. the housing top must be secured with screws attached to the base to prevent the two parts from falling apart at high temperatures). In other words, the conceptual model of product features can be defined 61 on the basis of the product features formulated as general design options or design alternatives. For example, the housing of a product can be made of different materials, such as plastic, metal or die-cast. With this approach of the present techniques, the conceptual model of product features of the majority of products can serve as a "building block" for new manufactured products or product variants, which can be built based on the design possibilities offered by the conceptual model of product features.

In the present disclosure, the knowledge graph of product features may include a plurality of product instances 40 associated with the respective concepts from the other plurality of concepts in the conceptual model of product features. In some examples, each product instance (41) may include data points (42, 43) of product attributes of the corresponding product in the plurality of products. In other words, a conceptual product feature model that represents common product features of the plurality of products (e.g., as general design options or design alternatives) can be instantiated by defining concrete product instances of the concepts contained in the conceptual product feature model that relate to a specific product (or a specific product cluster). In some examples, the product instances of specific products may be provided based on input from a product engineer (eg, using a user interface of a computer-implemented system) and stored in the knowledge graph of product features. Furthermore, the knowledge graph of product features can be implemented 62, for example, in a computer-implemented system together with the conceptual model of product features (eg, in a computer program running on the computer-implemented system). In some examples, the product feature knowledge graph may be implemented on the same computer-implemented system as the machine knowledge graph. In other examples, the product feature knowledge graph may be implemented on a computer-implemented system different than that on which the machine knowledge graph is implemented.

Furthermore, the step of “determining” 200 the one or more characteristics of the product to be manufactured in the machine knowledge graphs can include selecting 220 the one or more characteristics of the product to be manufactured based on the information contained in the knowledge graph of product characteristics. For example, based on the requirements specification for a new product or product variant, the product engineer may select certain options from the design alternatives or

Choose design options that are driven by the knowledge graph of

Product features are provided (e.g. through one or more concepts of the conceptual model of product features). Correspondingly, this selection can define the search criteria 63, 64 for the analysis of the manufacturability for the new product or the product variant. In some examples, this approach of the present techniques can be applied to pre-existing products or product variants. In other examples, the selection can be used for new products or product variants. For example, product engineers can verify manufacturability of new products early in the design phase (e.g., right after customer requirements are specified) using the techniques of the present disclosure. In addition, as discussed below, the present techniques may calculate changes made to product features and/or Machine skills need to be made to manufacture the new product.

In the present disclosure, the knowledge graph of product features can be configured to translate the selected one or more features of the product to be manufactured into the machine knowledge graph 31A, 32A. In some examples, the product engineer need not provide the full specification of all design alternatives for a new product. For example, if the product engineer B. does not select a specific design alternative for a specific product feature, this means that during the manufacturability analysis all design alternatives provided for this feature of the new product can be considered, which are stored in the knowledge graph of product features. That is, a set of product variants (e.g., given by the plurality of product instances) can be selected for the manufacturability analysis from the knowledge graph of product features, rather than a fully specified product where design alternatives have been specified for each feature of the product. In this context, for example in FIG. 2 the features relating to a size and a material of the product to be manufactured are shown as possible selected features. Furthermore, the machine knowledge graph of the present techniques can be configured to receive transmitted characteristics of the product to be manufactured.

In the present techniques, determining 300 the manufacturability of the product may further comprise, for each feature of the one or more selected features of the product to be manufactured, searching 310 for one or more machine instances from the plurality of machine instances whose data points of machine capabilities of the corresponding Machine from the plurality of machines provide manufacturability of the respective selected feature. As mentioned above, the machines can belong to the same group of machines or to different groups of machines, which can be located in the same industrial plant or in different industrial plants. The "search" step 310 of the method is performed using the machine knowledge graph. In some examples, the "determine" step 300 may include searching, for each feature from the one or more selected features of the product to be manufactured, after all machine instances from the plurality of machine instances whose data points of machine capabilities of the corresponding machine from the plurality of machines provide the manufacturability of the respective selected feature.

In the present disclosure, searching may include identifying 320, using the machine knowledge graph, one or more matches (e.g., exact matches) between the selected one or more features of the product to be manufactured and the machine instances contained in the machine knowledge graph. In some examples, identifying the one or more matches may be performed by comparing the selected one or more characteristics of the product to be manufactured and the machine capability data points contained in the machine instances. Furthermore, the method may include designating 330 one or more corresponding machines for each characteristic of the selected one or more characteristics of the product to be manufactured for which the match is identified. The product to be manufactured (e.g. a new product to be manufactured) can e.g. B. require a housing material made of plastic. In some examples, one or more machines about which information is contained in the machine knowledge graph may be capable of processing plastic housings. In this case an exact match can be identified, i. H. the one or more machines or all machines possessing that machine capability may be designated (or selected) (e.g., by executing the computer program mentioned above).

In the techniques of the present disclosure, searching may include identifying 340, using the machine knowledge graph, one or more value-based matches between the selected one or more features of the product to be manufactured and the machine instances contained in the machine knowledge graph. In this context, value-based matches are those matches that relate to features of the product that are characterized by one or more values. In some examples, identifying the one or more value-based Matching is performed by comparing the one or more selected characteristics of the product to be manufactured and the machine capability data points contained in the machine instances. Furthermore, identifying may further include calculating 350 the one or more value-based matches. In some examples, the calculating may be performed based on rules that determine when one of a plurality of machine capabilities can process a feature of the selected one or more features of the product to be manufactured. The selected characteristics of the product to be manufactured can e.g. B. define a product size in terms of length, width and height in millimeters. In some examples, based on the machine capability "Max Capability," a match may be identified by calculating (e.g., by executing the computer program mentioned above) if the product size in terms of length, width, and height is less than or equal to the is the maximum work area that the machine type or machine can handle. In a next step, the method may include designating 360 a corresponding machine or machines for each characteristic of the selected one or more characteristics of the product to be manufactured for which the value-based match is identified.

The techniques of the present disclosure may further include selecting 370 an identified machine or an identified plurality of machines from the plurality of machines that are appropriate for the manufacturability of the product to be manufactured for which the match or the value-based match with each feature of the one or the plurality of selected characteristics of the product to be manufactured. In this way, one or more machine capabilities can be determined from a plurality of machine capabilities that are compatible with each corresponding feature. Furthermore, a proportion of the selected machines may be determined 380 using the machine knowledge graph that are suitable for the manufacturability of the product to be manufactured. In some examples, the selected machines may be selected from a total number of machines in the plurality of machines present at an industrial facility. In addition, determining the proportion of selected machines for each industrial plant be carried out, which contains at least one selected machine that is suitable for the manufacturability of the product to be manufactured.

The next step of the present disclosure may further include determining 390, using the machine knowledge graph, one or more machines from the plurality of machines for which there is no match for one or more characteristics from the selected one or more characteristics of the product to be manufactured is identified. Additionally or alternatively, the method may further comprise determining, using the machine knowledge graph, one or more machines from the plurality of machines for which no value-based match is identified for one or more features from the selected one or more features of the product to be manufactured becomes. In some examples, the method may further include determining, using the machine knowledge graph, one or more features from the selected one or more features of the product to be manufactured for which no machine is identified from the plurality of machines with corresponding matches. Thereby, one or more machine capabilities can be determined from a plurality of machine capabilities that are incompatible with the one or more features of the product to be manufactured. Furthermore, this may further comprise determining one or more machines from the plurality of machines for which the match or the value-based match to each other characteristic of the selected one or more characteristics of the product to be manufactured is identified for which the determined one or the does not contain several features for which no machine is identified from the plurality of machines. For example, the information about mismatched machine capabilities can be valuable for the product or manufacturing engineers, since this information explains in detail why a certain product or a product variant cannot be manufactured on a certain machine or a certain type of machine.

The machine knowledge graph of the present techniques can be further configured to output information about a corresponding result of the searching step. In some examples, the output of the result can do that Providing information regarding the outcome to one

manufacturing engineer include. In addition, the output of the result can provide information regarding the result to a

include product engineer. Outputting the result may also include displaying information on a graphical user interface. Alternatively or additionally, the information regarding the outcome may be provided to a remote device or entity. In addition, the information regarding the result can be sent to a user.

Another step of the present disclosure may include selecting 400 one or more machines of the industrial facility on which to manufacture the product. In addition, the selection can be made on the basis of the information provided by the machine knowledge graph. In one example, the industrial facility may be classified 410 as suitable for the manufacturability of the product to be manufactured if the proportion of the selected machines that are present on the industrial facility and are suitable for the manufacturability of the product to be manufactured exceeds a predefined threshold. Otherwise, the industrial plant can be classified as unsuitable for the manufacturability of the product to be manufactured. So e.g. B. Those industrial plants are excluded from the manufacture of the product to be manufactured (e.g. by the product engineer) that have a percentage of the selected machines of less than, for example, 95% or less than 70% or less than 45%.

The techniques of the present disclosure may include modifying the one or more features from the selected one or more features of the product to be manufactured for which no machine is identified from the plurality of machines with the corresponding matches to enable the to be manufactured product is manufactured on at least one machine from the plurality of machines, wherein the modifying of the one or more features is based on the information provided by the machine knowledge graph. if e.g. B. If none of the machines are capable of producing the new product or product variant (e.g. if not all features of the product to be produced can be produced), the product engineer can produce them change “problematic” features 65 so that the new product can be manufactured. For example, if the new product variant requires a height of 240mm and the machine out of the majority of machines can only process products with a maximum height of 235mm, the product engineer can use this information provided by the machine knowledge graph, which, among other things, enables the calculation of the one or more value-based matches as disclosed above. In this example, the product engineer can check if the product can be redesigned so that the height of the product is less than 235mm.

Alternatively or additionally, if no machine is identified from the plurality of machines with the corresponding matches for the one or more characteristics, the method may comprise modifying one or more of a plurality of machine capabilities to enable the product to be manufactured at least one machine of the plurality of machines is manufactured. As mentioned above, modifying the one or more machine capabilities can be done based on the information provided by the machine knowledge graph. For example, if none of the machines in the plurality of machines are capable of manufacturing the new product or product variant (e.g. if not all features of the product to be manufactured cannot be manufactured), the manufacturing engineer can use the information provided by the machine knowledge graph in Use reference to the machine capabilities that are not compatible with the new product. For example, the manufacturing engineer may either reconfigure or modify 55 one or more machines so that the product can be manufactured.

The techniques of the present disclosure may include manufacturing products using the information provided by the machine knowledge graph.

A second general aspect of the present disclosure relates to a computer program configured to carry out the computer-implemented method according to the first general aspect. The present disclosure also relates to a computer-readable medium and signals storing or encoding the computer program of the present disclosure. A third general aspect of the present disclosure relates to a computer-implemented system for analyzing a manufacturability of products, configured to execute the computer program of the second general aspect. Furthermore, the computer-implemented system may be further configured to generate and/or use the machine knowledge graph. Additionally, the computer-implemented system may be further configured to generate and/or use the knowledge graph of product features. In addition, the computer-implemented system can be configured to manufacture products using the information provided by the machine knowledge graph

Control information of the first aspect. The computer-implemented system may include at least one processor, at least one memory (which may contain programs that, when executed, perform the methods of the present disclosure), and at least one interface for inputs and outputs.

Claims

PATENT CLAIMS

A computer-implemented method for analyzing a manufacturability of products, the method comprising: providing (100) a computer-implemented machine knowledge graph (1) containing information about a plurality of machines, the machine knowledge graph comprising : a conceptual model (10) comprising a plurality of concepts (11, 12), each concept mapping a corresponding one of a plurality of machine capabilities, and wherein the concepts are provided with attributes (13-15) and their respective relationships are defined among themselves; a plurality of machine instances (20) associated with the respective concepts in the conceptual model, each machine instance (21) having data points of machine capabilities (22-24) of the corresponding one of the plurality of machines; providing (200) one or more characteristics of a product to be manufactured in the machine knowledge graphs; determining (300) manufacturability of a product based on the information contained in the machine knowledge graph.

The method of claim 1, wherein providing (200) the one or more features of the product to be manufactured in the machine knowledge graph further comprises providing (210) a knowledge graph of product features (2) containing information about a plurality of products , wherein the knowledge graph of product features comprises: a conceptual model of product features (30) comprising another plurality of concepts (31, 32), each concept from the other plurality of concepts mapping a corresponding one of a plurality of product features, wherein the concepts from the other plurality of concepts are provided with attributes and their respective interrelationships are defined; a plurality of product instances (40) associated with respective ones of the other plurality of concepts in the conceptual model of product features, each product instance (41) having data points (42, 43) of product features of the corresponding product from the plurality of products.

The method of claim 2, wherein providing the one or more characteristics of the product to be manufactured in the machine knowledge graph comprises selecting (220) the one or more characteristics of the product to be manufactured based on the information contained in the knowledge graph of product characteristics.

The method of claim 3, wherein the knowledge graph of product features is configured to transfer the one or more selected features of the product to be manufactured into the machine knowledge graph (31A, 32A), and wherein the machine knowledge graph is configured to transfer Receiving characteristics of the product to be manufactured.

The method of claim 3 or 4, wherein determining (300) manufacturability of the product further, for each feature of the selected one or more features of the product to be manufactured, searching (310) for one or more machine instances of the plurality of Machine instances whose data points of machine capabilities of the corresponding machine from the plurality of machines provide manufacturability of the respective selected feature, wherein the searching is performed using the machine knowledge graph.

The method of claim 5, wherein the searching comprises identifying (320), using the machine knowledge graph, one or more matches between the one or more selected features of the product to be manufactured and the machine instances contained in the machine knowledge graph, optionally wherein identifying the one or more matches is performed by comparing the one or more selected characteristics of the product to be manufactured and the machine capability data points contained in the machine instances, further optionally the method further designating (330) one or more corresponding machines for each Feature from the one or more selected features of the product to be manufactured for which the match is identified.

The method of claim 5 or 6, wherein the searching comprises identifying (340), using the machine knowledge graph, one or more value-based matches between the one or more selected features of the product to be manufactured and the machine instances contained in the machine knowledge graph , optionally wherein identifying the one or more value-based matches is performed by comparing the selected one or more characteristics of the product to be manufactured and the machine capability data points contained in the machine instances.

The method of claim 7, wherein the identifying further comprises calculating (350) the one or more value-based matches, the calculating being performed based on rules that determine when a machine capability from a plurality of machine capabilities has a characteristic from the one or the selected plurality of characteristics of the product to be manufactured, optionally wherein the method further comprises designating (360) one or more corresponding machines for each characteristic of the selected one or more characteristics of the product to be manufactured for which the value-based match is identified.

9. The method according to any one of the preceding claims 6 to 8, further comprising selecting (370) an identified machine or an identified plurality of machines from the plurality of machines that are suitable for the manufacturability of the product to be manufactured for which the correspondence or the value-based correspondence with each feature is identified from the one or more selected features of the product to be manufactured, and thus determining one or more machine capabilities from a plurality of machine capabilities that are compatible with each corresponding feature, optionally wherein the method further comprises determining ( 380), using the machine knowledge graph, a proportion of the selected machines that are suitable for the manufacturability of the product to be manufactured, the selected machines being selected from a total number of machines from the plurality of machines located on an In industrial plant are available.

10. The method according to any one of the preceding claims 6 to 9, further comprising determining, using the machine knowledge graph, one or more features from the one or more selected features of the product to be manufactured for which no machine of the plurality of machines with corresponding matches are identified, and thus determining one or more machine capabilities from a plurality of machine capabilities that are incompatible with the one or more characteristics of the product to be manufactured.

11. The method of claim 10, further comprising modifying the one or more characteristics from the selected one or more characteristics of the product to be manufactured for which no machine is identified from the plurality of machines with the corresponding matches to enable the product to be manufactured is manufactured on at least one machine from the plurality of machines, wherein the one or more features are modified on the basis of the information provided by the machine knowledge graph.

12. The method of claim 10 or 11, further comprising modifying one or more of a plurality of machine capabilities when no machine of the plurality of machines is identified with the corresponding matches for the one or more characteristics to enable the product to be manufactured is manufactured on at least one machine from the plurality of machines, the modification of the one or more machine capabilities being based on the information provided by the machine knowledge graph.

13. A computer program designed to carry out the computer-implemented method according to any one of the preceding claims.

14. A computer-implemented system for analyzing a manufacturability of products, adapted to execute the computer program according to claim 13.